Aim: The aim of the research article is to improve the accuracy of movie recommendation systems using a novel Random forest (RF) algorithm in comparison with a Naive Bayes (NB) algorithm. Materials and methods: The dataset used in this paper was collected from the Movie lens database. The sample size for the movie recommendation system was sample 20 (Group 1 = 10 and Group 2 = 10) and the calculation was performed utilizing G-power 0.8 with alpha and beta qualities of 0.05 and 0.2 with a confidence interval of 95%. The movie recommendation system is performed by the Random forest (RF) classifier with a number of samples (N=10) and Naive Bayes (NB) model with a number of samples (N=10). Results: The Random forest (RF) classifier has a 94.30 percent higher accuracy rate when compared to the accuracy rate of the Naive Bayes (NB) model, which is 82.56 percent. The study has a significance value of p=0.037. Conclusion: The Random forest (RF) classifier provides better outcomes in terms of accuracy rate when compared to the Naive Bayes (NB) model for movie recommendation systems.